Increased hepatic PDGF-AA signaling mediates liver insulin ... · 5/2/2018 · 1 Increased hepatic...

1

Increased hepatic PDGF-AA signaling mediates liver insulin resistance in obesity

associated type 2 diabetes

Amar Abderrahmani1,2*, Loïc Yengo

1*, Robert Caiazzo

3*, Mickaël Canouil

1*, Stéphane

Cauchi1, Violeta Raverdy

3, Valérie Plaisance

1, Valérie Pawlowski

1, Stéphane Lobbens

1,

Julie Maillet1, Laure Rolland

1, Raphael Boutry

1, Gurvan Queniat

1, Maxime Kwapich

1,

Mathie Tenenbaum1, Julien Bricambert

1, Sophie Saussenthaler

4, Elodie Anthony

5,

Pooja Jha6, Julien Derop

1, Olivier Sand

1, Iandry Rabearivelo

1, Audrey Leloire

1, Marie

Pigeyre3, Martine Daujat-Chavanieu

7, Sabine Gerbal-Chaloin

7, Tasnim Dayeh

8,

Guillaume Lassailly9, Philippe Mathurin

9, Bart Staels

10, Johan Auwerx

6, Annette

Schürmann4, Catherine Postic

5, Clemens Schafmayer

11, Jochen Hampe

12, Amélie

Bonnefond1,2, François Pattou

3*, Philippe Froguel

1,2*

1Univ. Lille, CNRS, Institut Pasteur de Lille, UMR 8199 - EGID, F-59000 Lille, France;

2Department of Medicine, Section of Genomics and Common Disease, Imperial College

London, UK; 3Univ. Lille, Inserm, CHU Lille, U1190 - EGID, F-59000 Lille, France;

4Department of Experimental Diabetology, German Institute of Human Nutrition Potsdam-

Rehbrüecke, Nuthetal and German Center for Diabetes Research (DZD), München-

Neuherberg, Germany. 5Inserm, U1016, Institut Cochin, Paris, France CNRS UMR 8104,

Paris, France Université Paris Descartes, Sorbonne Paris Cité, Paris, France. 6Laboratory of

Integrative and Systems Physiology, École Polytechnique Fédérale de Lausanne, 1015

Lausanne, Switzerland. 7INSERM U1183, Univ. Montpellier, UMR 1183, Institute for

Regenerative Medicine and Biotherapy, CHU Montpellier, France; 8Department of clinical

science; Skane University Hospital Malmö, Malmö, Sweden; 9Univ. Lille, Inserm, CHU Lille,

U995 - LIRIC - Lille Inflammation Research International Center, F-59000 Lille, France;

Page 2 of 66Diabetes

Diabetes Publish Ahead of Print, published online May 4, 2018

2

10Univ. Lille, Inserm, CHU Lille, Institut Pasteur de Lille, U1011- EGID, F-59000 Lille,

France; 11 Department of Visceral and Thoracic Surgery, University Hospital Schleswig-

Holstein, Kiel, Germany; 12Medical Department 1, Technische Universität Dresden (TU

Dresden), Dresden, Germany.

*These authors equally contributed to the study

Corresponding authors:

Philippe Froguel, MD, PhD, [email protected], and Amar Abderrahmani, PhD,

[email protected]

Abstract

In type 2 diabetes (T2D), hepatic insulin resistance is strongly associated with non-

alcoholic fatty liver disease (NAFLD). In this study, we hypothesized that DNA methylome

of livers from patients with T2D, when compared to livers of individuals with normal plasma

glucose levels, can unveil some mechanism of hepatic insulin resistance that could link to

NAFLD. Using DNA methylome and transcriptome analyses of livers from obese individuals,

we found that both hypomethylation at a CpG site in PDGFA (encoding platelet derived

growth factor alpha) and PDGFA overexpression are associated with increased T2D risk,

hyperinsulinemia, increased insulin resistance and increased steatohepatitis risk. Both genetic

risk score studies and human cell modeling pointed to a causative impact of high insulin

levels on PDGFA CpG site hypomethylation, PDGFA overexpression, and increased PDGF-

AA secretion from liver. We found that PDGF-AA secretion further stimulates its own

expression through protein kinase C activity and contributes to insulin resistance through

decreased expression of both insulin receptor substrate 1 and of insulin receptor. Importantly,

Page 3 of 66 Diabetes

3

hepatocyte insulin sensitivity can be restored by PDGF-AA blocking antibodies, PDGF

receptor inhibitors and by metformin opening therapeutic avenues. Therefore, in the liver of

obese patients with T2D, the increased PDGF-AA signaling contributes to insulin resistance,

opening new therapeutic avenues against T2D and possibly NAFLD.

Keywords: PDGFA, Type 2 diabetes, Obesity, Liver, Epigenetics

Abbreviations

ABOS: Atlas Biologique de l'Obésité Sévère

BMI: Body Mass Index

BMIQ: Beta-MIxture Quantile normalization

DMRs: Differentially Methylated Regions

FDR: False Discovery Rate

GRS: Genetic Risk Score

GWAS: Genome-wide association studies

HCC: Hepatocarcinoma

IHH: immortalized human hepatocytes (IHH)

INSR: Insulin Receptor

IRS1: insulin receptor substrate 1 (IRS1)

NAFLD: Non-Alcoholic Fatty Liver Disease


4

NASH: Non-Alcoholic Steato-Hepatitis

PCA: Principal Component Analysis

PDGFA: Platelet Derived Growth Factor A

PMA: Phorbol 12-myristate 13-actetate

PKC: Protein Kinase C

SNPs: Single Nucleotide Polymorphisms

STKs: Serine/threonine protein kinases


5

In type 2 diabetes (T2D), hepatic insulin resistance is a major contributor of fasting

and postprandial hyperglycemia. Although intrahepatic increased lipids and chronic elevated

plasma insulin have been incriminated, the intracellular molecular mechanism that account for

the impaired insulin signaling in livers of patients with T2D is still incompletely understood.

Moreover, in T2D, hepatic insulin resistance is strongly associated with non-alcoholic fatty

liver disease (NAFLD) (1), suggesting that the mechanisms leading to hepatic insulin

resistance contribute to the development of NAFLD, and conversely. Genome-wide

association studies (GWAS) and related metabolic traits have identified many loci associated

with the risk of T2D (2). However, these loci only explain 15% of T2D inheritance, and

GWAS have opened limited insights into the pathophysiology of T2D including hepatic

insulin resistance (2). Several DNA methylome-wide association studies have identified

candidate genes possibly involved in metabolic dysfunction of the adipose tissue, skeletal

muscle and liver in obesity and T2D (3–7). Further, epigenetic analyses of livers from obese

individuals with diabetes have enabled the identification of altered expression in genes

involved in glucose and lipid metabolism (3). So far, the causality of such dysregulation in the

development of liver insulin resistance has not been determined.

In this study, we hypothesized that analysis of the DNA methylome of livers from

patients with T2D, when compared to livers of individuals with normal plasma glucose levels,

can unveil key players of hepatic insulin resistance in response to a diabetogenic environment.

Our DNA methylome- and transcriptome-wide association analyses for T2D in liver samples

from obese subjects identified a reduced methylation of a CpG site within the platelet-derived

growth factor A gene (PDGFA), which was correlated with an increase in PDGFA mRNA.

PDGFA encodes a protein forming a PDGF-AA homodimer, which is known as a liver

fibrosis factor when overexpressed in the liver (8). We found that the rise of liver PDGFA is

not only associated with increased T2D risk, but also with increased steatohepatitis (NASH)


6

risk, elevated fasting plasma insulin and insulin resistance. The increased PDGFA expression

and PDGF-AA protein levels were reproduced in human hepatocytes made insulin-resistant

by long-term insulin incubation, and were present in the liver of insulin-resistant rodents.

Furthermore, using human hepatocytes, we have demonstrated that PDGF-AA overexpression

perpetuated hepatocyte insulin resistance in an autocrine feed-forward loop mechanism,

providing novel insights into the mechanism possibly linking hepatic insulin resistance and

NAFLD in diabetes associated obesity.

Research Design and Methods

Discovery study. Liver biopsies were collected by surgery of 192 French obese subjects.

Subjects included in the discovery study were participants of the ABOS (“Atlas Biologique de

l'Obésité Sévère”) cohort (ClinicalGov NCT01129297) including 750 morbidly obese subjects

whose several tissues were collected during bariatric surgery (9). All subjects were unrelated,

women, above 35 years of age, of European origin verified by Principal Component Analysis

(PCA) using SNPs on the Metabochip array, non-smoker, non-drinker, without any history of

hepatitis, and without indications of liver damage in serological analysis (normal ranges of

aspartate aminotransferase and alanine aminotransferase). However, we observed a significant

increase in aspartate aminotransferase and alanine aminotransferase concentrations in patients

with T2D when compared to controls (table S1). Overall, 96 T2D cases and 96

normoglycemic participants were selected. Normoglycemia and T2D were defined using the

World Health Organization/International Diabetes Federation 2006 criteria (Normoglycemia:

fasting plasma glucose < 6.1 mmol/l or 2-h plasma glucose < 7.8 mmol/l; T2D: fasting plasma

glucose ≥ 7 mmol/l or 2-h plasma glucose ≥ 11 mmol/l). Each participant of the ABOS cohort

signed an informed consent. For calculation of intermediate metabolic traits (HOMA2-IR and


7

HOMA2-B indexes), see the supplemental information. All procedures were approved by

local ethics committees. The main clinical characteristics were presented in Table S1.

Replication study. The replication study was based on in silico data of liver samples

analyzed by the Infinium HumanMethylation450 BeadChip, as previously reported (10).

Clinical characteristics were reported in Table S2. All patients provided written, informed

consent. The study protocol was approved by the institutional review board

(‘‘Ethikkommission der Medizinischen Fakultät der Universität Kiel,’’ D425/07, A111/99)

before the beginning of the study. Liver samples were obtained percutaneously from subjects

undergoing liver biopsy for suspected NAFLD or intraoperatively for assessment of liver

histology. Normal control samples were recruited from samples obtained for exclusion of

liver malignancy during major oncological surgery. Moreover, patients with evidence of viral

hepatitis, hemochromatosis, or alcohol consumption greater than 20 g/day for women and 30

g/day for men were excluded. None of the normal control subjects underwent preoperative

chemotherapy, and liver histology demonstrated absence of both cirrhosis and malignancy. A

percutaneous follow-up biopsy was obtained in consenting bariatric patients five to nine

months after surgery. Biopsies were immediately frozen in liquid nitrogen, ensuring an ex

vivo time of less than 40 seconds in all cases.

Genome-wide analysis of DNA methylation. The genome-wide analysis of DNA

methylation was performed using the Infinium HumanMethylation450 BeadChip (Illumina,

Inc., San Diego, CA, USA) which interrogates 482,421 CpG sites and 3,091 non-CpG sites

covering 21,231 RefSeq genes (11). We used 500 ng of DNA from liver tissue for bisulfate

conversion using the EZ DNA Methylation kit D5001 (Zymo Research, Orange, CA, USA)

according to the manufacturer’s instructions. For the details of procedure, see the

supplemental information.


8

SNP genotyping, ethnic characterization and genetic risk score. SNP genotyping was

performed with Metabochip DNA arrays (custom iSelect-Illumina genotyping arrays) using

the Illumina HiScan technology and GenomeStudio software (Illumina, San Diego, CA, USA)

(12). We selected SNPs with a call rate ≥ 95 % and with no departures from Hardy–Weinberg

equilibrium (p>10-4). A Principal Component Analysis (PCA) was performed in a combined

dataset involving the 192 patients plus 272 subjects from the publicly available HapMap

project database. For these 272 subjects (87 of European ancestries [HapMap CEU], 97 of

Asian ancestries [HapMap CHB] and 88 of African ancestries [HapMap YRI]) genotype calls

at the 106,470 SNPs present on the Metabochip were available. The first two components

were sufficient to discriminate ethnic origin (Fig. S1) and we observed that study participants

clustered well with HapMap samples of European ancestries. For details of analysis, see the

supplemental information.

Statistical Analyses. Statistical analysis and quality control were performed with R software

version 3.1.1. Raw data (IDAT file format) from Infinium HumanMethylation450 BeadChips

were imported into R using the minfi package (version 1.12.0 on Bioconductor) (13), then we

applied the preprocessing method from GenomeStudio software (Illumina) using the reverse

engineered function provided in the minfi package. Samples were excluded when less than 75

% of the markers had detection p-values below 10-16. Markers were ruled out when less than

95 % of the samples had detection p-values below 10-16. According to this strategy, no sample

was excluded and 70,314 markers (out of 485,512) were excluded. For correction for Infinium

HumanMethylation450 BeadChip design which includes two probe types (Type I and Type

II), a Beta-MIxture Quantile normalization (BMIQ) was performed (14). Moreover, we

checked for outliers using Principal Component Analysis (PCA) (flashpcaR package, version

1.6-2 on CRAN). At this stage, 416,693 markers and 192 samples were kept for further

analysis. To test the association between methylation level and diabetic status, we applied a


9

linear regression adjusted for age, BMI, steatosis (in percent), presence of NASH and fibrosis.

Results were corrected for multiple testing using a Bonferroni correction (p<10-7). The

association between DNA methylation and metabolic traits was analyzed using a linear

regression model, including normoglycemic samples adjusted for age and BMI. Quality

control was performed on the HumanHT-12 v4.0 Whole-Genome DASL HT Assay (Illumina)

data, according to the following criterion: probes were kept for further analysis when the

detection p-values provided by GenomeStudio software version 3.0 (Illumina) were below

five percent for all samples. A PCA was performed to identify samples with extreme

transcriptomic profiles. After the quality control just described, 18,412 probes matching

13,664 genes and 187 samples were kept and analyzed for differential expression between

T2D cases and controls, using linear regression. Methylation and expression data were tested

for correlation. Linear regression analysis was used for testing the association in cis- genes

within the 500 kb with the CpG methylation site with the T2D status as a covariate. To

account for multiple testing, we used five percent as a threshold for false discovery rate

(FDR).

We selected a subgroup of 24 samples among the 192 initial samples, including 12

normoglycemic and 12 T2D cases, to analyze DNA methylation in blood samples from the

same donors. The 24 samples were selected based on their expression and methylation

profiles using PCA to reduce the heterogeneity.

Cell Culture. Immortalized Human Hepatocytes (IHH) were obtained from primary human

hepatocytes that were transfected with a plasmid carrying the large T antigen SV40 (15). The

cells retained features of normal hepatocytes including albumin secretion, the multidrug

resistance (MDR) P-glycoprotein, active uptake of the bile salt taurocholate. triglyceride

(TG)-rich lipoproteins, apolipoprotein B (0.6 mg/ml per day) and apolipoprotein A-I secretion

(15). IHH cells (maintained at passages 35-45) were cultured in Williams E medium


10

(Invitrogen), containing 11 mM glucose and supplemented with 10 % fetal calf serum (FCS;

Eurobio), 100 U/ml penicillin, 100 µg/ml streptomycin, 20 mU/ml insulin (Sigma-Aldrich)

and 50 nM dexamethasone (Sigma-Aldrich) (16). For insulin pre-treatment, 106 cells were

cultured in 6-well plates in a Dulbecco’s Modified Eagle Medium (DMEM; Invitrogen) with

or without 100 nM human insulin (Novo Nordisk) supplemented with 5 mM Glucose, 2 %

FCS, 100 U/ml penicillin, 100 µg/ml streptomycin for 24 hours. For monitoring insulin

signaling, medium was removed and replaced by FCS- and phenol red-free DMEM medium

with or without 200 nM human insulin for one hour. Human hepatocytes were isolated from

liver lobectomies resected for medical reasons as described (17) in agreement with the ethics

procedures and adequate authorization.

RNA sequencing, Microarray mRNA expression analysis, qRT-PCR, Western Blotting,

Chemicals, ELISA, Glycogen measurement, Global Serine/Threonine kinases activity,

DNA/RNA preparation, Oil-Red staining, cell proliferation, apoptosis, intermediate

metabolic traits. See the supplemental experimental procedure in the supplemental

information


11

Results

Liver epigenetic modification in T2D

The liver DNA methylome was assessed in 96 age- and body mass index (BMI)-

matched obese women with T2D and 96 obese women with normal glucose levels (Table S1).

While we initially identified 381 differentially methylated regions (DMRs) in the liver from

obese patients with T2D compared to normal glucose obese patients (Fig. S2), we only

observed one genome-wide significant DMR (cg14496282 within PDGFA) in obese patients

with T2D after adjusting for liver steatosis, and NASH, in an attempt to control for

confounding effects (Table S1). The methylation at cg14496282 was associated with

decreased T2D risk (β = -15.6 %; p = 2.5×10-8; Fig. 1a and 1b). The average DNA

methylation at the cg14496282 was 41.3 % in patients with T2D and 60.3 % in controls,

which corresponds to a 1.46-fold decrease in the methylation level of the CpG site. We

checked the methylation, at this CpG site, by using mathematical deconvulation analyses (18),

for possible confounding effects due to differences in cell composition, and we still observed

consistent effects for T2D risk (β = -14.9 %; p = 6.9×10-7). We replicated this association in

the liver from 12 cases with T2D and 53 German control subjects (10), and found that T2D

risk was associated with decreased methylation level at cg14496282 site (β = -14.0 %; p =

0.01) (Table S2). A decreased methylation level in another CpG site within PDGFA has been

found in the liver from obese men with T2D compared to non-obese controls from other

cohort (3), supporting that PDGFA is a target for epigenetic modification in response to

obesity-associated diabetes

We next investigated whether the T2D-associated PDGFA cg14496282

hypomethylation was specific to the liver. We assessed the blood DNA methylome from 12

obese cases with T2D and 12 obese normal glucose controls presenting with extreme liver


12

methylation levels at cg14496282. We found a significant correlation between methylation

levels in blood and liver (r = 0.66; p = 6.61×10-4), and a slightly reduced methylation at the

cg14496282 site (β = -1.4 %; p = 0.01) in blood of subjects with T2D when compared to

controls. We also compared DNA methylation at cg14496282 in 43 liver and skeletal muscle

samples from 192 participants who were randomly selected from the ABOS cohort, but we

did not find any significant correlation (p > 0.05).

In the 192 obese liver samples, we next investigated cis-located genes (within 500 kb

around cg14496282) that were differentially expressed between T2D cases and controls, and

which mRNA expression correlated with DNA methylation at PDGFA cg14496282 site.

Using a false discovery rate threshold of 5 % for differential expression analysis and

methylation-expression correlation analysis, we identified that the methylation at cg14496282

is negatively associated with the expression of PDGFA in T2D cases and normal glucose

controls (p < 0.007; Table 1).

Reduced liver PDGFA expression is associated with lower hepatic fibrosis risk

In subjects with T2D and in normoglycemic controls, we found that PDGFA

cg14496282 methylation was significantly associated with decreased NASH risk (p < 0.05;

Table 1), while PDGFA expression in the liver was associated with increased NASH risk (p <

0.01; Table 1). Furthermore, in patients with T2D, PDGFA cg14496282 methylation was

significantly associated with decreased hepatic fibrosis, decreased alanine aminotransferase

levels and decreased aspartate aminotransferase levels (p < 0.05; Table 1), while PDGFA

expression in the liver was associated with increased hepatic fibrosis and increased liver

enzyme levels (p < 0.01; Table 1). These results were in line with previous studies that

showed that PDGFA cg14496282 hypomethylation is associated with increased PDGFA liver


13

expression in advanced versus mild human NAFLD (19,20). PDGFA encodes a dimer

disulfide-linked polypeptide (PDGF-AA) that plays a crucial role in organogenesis and

cirrhotic liver regeneration (21,22). Overexpression of PDGF-AA in mice liver causes

spontaneous liver fibrosis (8). Moreover, activation of PDGF receptor signaling stimulates

hepatic stellate cells and thereby promotes liver fibrosis (23–25).

Increased liver PDGFA expression is associated with hyperinsulinemia and insulin

resistance

In obese subjects with normal glucose levels, we next found that PDGFA cg14496282

methylation is significantly associated with decreased fasting serum insulin levels and

decreased insulin resistance as modeled by the homeostasis model assessment index HOMA2-

IR (β = -1.45×10-3, p = 2.32×10-3; and β = -0.10, p = 4.93×10-3, respectively; Table 1). In

contrast, PDGFA liver expression was significantly associated with increased fasting serum

insulin levels and increased insulin resistance (β = 6.83×10-3, p = 9.49×10-3; and β = 0.53, p =

7.47×10-3, respectively; Table 1).

Subsequently, we calculated a genetic risk score (GRS) as the sum of alleles

increasing fasting insulin levels over 19 GWAS-identified single nucleotide polymorphisms

(SNPs) (26), and found that this GRS is associated with decreased DNA methylation at

cg14496282 (β = -1.05 % per allele; p = 4×10-3; Table S3). This association remained

significant when we analyzed T2D cases and controls separately (and then meta-analyzed) or

when we adjusted for BMI, high-density lipoprotein (HDL) cholesterol or triglycerides; these

traits having a genetic overlap with fasting insulin (26). These results strongly suggested that

hyperinsulinemia (and associated insulin resistance) contributes to decreased DNA

methylation of PDGFA cg14496282 and consequently to the increase in the PDGFA


14

expression. In contrast, the GRS including 24 SNPs associated with fasting glucose, the GRS

including 65 SNPs associated with T2D and the GRS including 97 SNPs associated with BMI

were not associated with cg14496282 methylation (Table S3). These results suggest that

hyperglycemia and obesity per se are not involved in the modulation of PDGFA methylation

and expression.

The in vivo association between liver PDGFA overexpression and insulin resistance

was supported by the data obtained from different mice models of insulin resistance

associated with obesity. In the liver from C57BL/6J (B6) mice that are susceptible to diet-

induced obesity (27), we found that Pdgfa expression is increased by 46 %, as compared with

control mice (i.e. that do not respond to a high-fat diet) (Fig. 1c). Similarly, we found that

liver Pdgfa expression is increased in insulin resistant BXD mice fed a high-fat diet for 21

weeks when compared to control mice (Fig. 1d). However, the cg14496282 CpG site is not

conserved in mice (28), suggesting that different epigenetic mechanisms rely on the rise of

Pdgfa/PDGFA in insulin-resistant hepatocytes in mice and humans.

Increased PDGFA expression and secretion from insulin-resistant human hepatocytes

The association of increased liver PDGFA expression with systemic insulin resistance

in obese subjects suggests that PDGFA overexpression plays a role in liver insulin resistance,

and thereby in T2D development. Chronic hyperinsulinemia indeed induces liver insulin

resistance (29) and PDGFA has an autocrine function on hepatocytes (21). In this context, the

exposure of mouse embryo cells to PDGF-AA inhibits insulin signaling (30). Therefore, we

hypothesized that the obesity-associated increased PDGFA expression contributes to mediate

the deleterious effects of chronic hyperinsulinemia on hepatic insulin resistance. To assess

this hypothesis, we established an in vitro model of insulin-resistant human hepatocytes


15

caused by hyperinsulinemia. To do so, we used the immortalized human hepatocytes (IHH).

IHH cells are indeed equipped will the functional machinery for glucose metabolism (16) and

they secrete PDGF-AA homodimer at comparable levels with primary human hepatocytes

(Fig. 2a). Exposure of IHH cells to insulin for 16 h or 24 h hampered insulin-induced

phosphorylation of AKT serine/threonine kinase at residue serine 473 (Fig. 2b). In line with

AKT activation pivotal role in glycogen synthesis (31), we found reduced insulin-induced

glycogen production in IHH cells exposed to insulin for 16 h and 24 h (Fig. 2c). The defective

insulin signaling by insulin was accompanied by a rise in PDGFA mRNA, and in abundance

and secretion of the encoded protein PDGF-AA homodimer from IHH cells (Fig. 2d-f). In

addition, the insulin-induced increase in PDGFA expression was associated with the reduced

cg14496282 CpG methylation level in IHH cells (Fig. 2g), suggesting that insulin resistance

induced by hyperinsulinemia accounts for the rise of PDGFA in hepatocytes of obese

individuals with diabetes. As these results may be cell line dependent, we measured AKT

phosphorylation in liver hepatocellular HepG2 cells exposed to insulin for 24 h and retrieved

similar results (Fig. 2h). In HepG2 cells, the expression of PDGFA was also significantly

increased after long-term insulin incubation (Fig. 2i). The increase in PDGFA mRNA seemed

specific to insulin exposure as neither cell proliferation (Fig. S3) nor the intracellular neutral

lipid levels was modified by insulin treatment (Fig. S4). On the other hand, palmitate

exposure did not change PDGFA expression in IHH cells (Fig. S4).

PDGF-AA contributes to insulin resistance induced by insulin

Our data suggest that it is hyperinsulinemia, and not the excess of fatty acids influx,

which induces PDGF-AA secretion in the liver of obese individuals with T2D. We then

hypothesized that PDGF-AA directly causes insulin resistance in human hepatocytes. In line


16

with this hypothesis, we found that the culture of IHH cells with a human PDGF-AA

recombinant inhibits insulin-induced AKT activation (Fig. 3a). In contrast, the incubation of

IHH cells with anti-PDGF-AA blocking antibodies reversed the deleterious effect of insulin

long term exposure on AKT phosphorylation (Fig. 3b). We then investigated the mechanism

whereby PDGF-AA inhibits AKT activation. Human hepatocytes express PDGF receptors

(PDGFR) including PDGFRα and PDGFRβ that both bind PDGF-AA (8). We tested the role

of PDGFR signaling using the PDGFR inhibitor Ki11502 (32). Pre-treatment of IHH cells

with Ki11502 efficiently antagonized the negative effect of insulin on AKT phosphorylation

mimicking results obtained with PDGF-AA blocking antibodies (Fig. 3c). PDGFR blockade

by Ki11502 increased the ability of insulin to stimulate glycogen synthesis (Fig. 3d).

To further dissect the signaling pathways by which both insulin and PDGF-AA impair

AKT and insulin action, we performed a global measurement of serine/threonine protein

kinases (STKs) using STK PamGene arrays consisting of 140 immobilized serine/threonine-

containing peptides that are targets of most known kinases (33). We looked for differential

STK activity between control and IHH cells cultured with insulin for 24 h. Peptides whose

phosphorylation varied significantly between the two conditions were indicative of

differential specific STK activities. This unbiased kinase analyses underscored significant

differences in protein kinases C (PKCƟ and PKCε) activities (Fig. 4a). The activation of

these two PKCs hampers insulin signaling in response to chronic hyperlipidemia (34–36).

These two kinases are also known to phosphorylate the insulin receptor substrate 1 (IRS1) and

the insulin receptor (INSR) on serine residues, that impairs the association of INSR with IRS

proteins, leading to the blockade of AKT activation and of the downstream signaling

pathways (35,36). Therefore, we treated IHH cells with phorbol 12-myristate 13-actetate

(PMA), a potent activator of PKCs, and retrieved AKT inhibition (Fig. S5). PKCƟ and PKCε

kinase activities are linked to their phosphorylation at Serine 676 and Serine 729, respectively


17

(37,38). In IHH cells cultured with insulin for 16 h or 24 h, we found a striking

phosphorylation of the two PKCs, which coincided with the decreased AKT phosphorylation

(Fig. 4b). The effect of insulin on the phosphorylation of the two kinases is likely to rely on

PDGF-AA, as the PKCƟ and PKCε were directly activated by PDGF-AA (Fig. 4c).

Activation of PKCε decreases INSR abundance (34). In line with this result, we found in IHH

cells cultured with insulin and PDGF-AA that an impaired INSR content is closely linked to

reduced INSR tyrosine phosphorylation at residue Y972 (Fig. 4d). In addition, the PKCε-

mediated T1376 phosphorylation which inactivates the INSR, was increased in response to

insulin or PDGF-AA (Fig. 4e).

To gain further insights into the intracellular mechanism through which PDGF-AA

alters insulin signaling in hepatocytes, we performed RNA sequencing of IHH cells treated or

not with insulin for 24 h. We found a profound dysregulation of expression of genes involved

in both carbohydrate metabolism, inflammatory and insulin signaling pathways in response to

insulin. Indeed, when we grew a network based on PDGFA through Ingenuity Pathway

Analysis (IPA), we found a significant increase in the expression of genes of the VEGF and

PDGF families, including as expected PDGFA (log2 Fold Change = 0.80; p = 1.1×10-11) (Fig.

S6 and Table S5). Subsequently, we analyzed the diseases and/or functions highlighted by

the insulin-evoked deregulated expressed genes in IHH cells. Among the significant outputs,

we found a network related to the metabolism of carbohydrates that includes PDGFA (p =

1.2×10-6; Fig. S7, Table S5). We also identified in IHH cells cultured with insulin for 24 h a

decrease in the expression of the insulin receptor substrate 1 (IRS1) gene (Fig. 5a, Fig. S6

and Table S6). The decreased IRS1 expression by insulin was confirmed by western blotting

(Fig. 5b) and mimicked by PDGF-AA (Fig. 5c). Silencing of IRS1 expression using small

interfering RNAs confirmed that the IRS1 abundance is critical for AKT activation in IHH

cells in response to insulin (Fig. 5d). Defective IRS1 level can therefore account for the


18

impaired insulin signaling caused by PDGF-AA in insulin resistant hepatocytes. The decrease

of IRS1 mRNA and protein level is mediated by PKC activity since its inhibition by the

PKCƟ and PKCε, inhibitor sotrastaurin (39) prevented the reduction of IRS1 caused by

hyperinsulinemia (Fig. 5e and 5f), and in contrast, PKC activation by (PMA) mimicked the

effect of insulin (Fig. 5g and 5h).

Altogether, our data suggest a role for hepatocyte PDGF-AA in promoting further

liver insulin resistance via the decrease of INSR and IRS1, and the activation of both PKCƟ

and PKCε.

Autocrine regulation of PDGF-AA on its expression and the beneficial effects of

metformin

PDGF-AA stimulates its own expression in the liver (40). It suggests that in insulin

resistant hepatocytes PDGF-AA amplifies its secretion, thereby perpetuating insulin

resistance. We found that culture of IHH cells with PDGF-AA stimulated PDGFA expression

(Fig. 6a) and PDGF-AA secretion (Fig. S6). The effect of PDGF-AA on its expression was

mediated by PDGFR as the PDGFR tyrosine kinase inhibitor ki11502 prevented the rise of

PDGFA mRNA of cells exposed to either insulin or to human PDGF-AA (Fig. 6a and 6b).

The overexpression of PDGFA by PDGF-AA may require PKC activation since the PMA

mimicked both insulin and PDGF-AA effects on the PDGFA mRNA (Fig. 6d) and inversely,

the inhibitor sotrastaurin, alleviated the rise of PDGFA (Fig. 6e). Altogether, it is likely that in

T2D, hyperinsulinemia-induced PDGF-AA aggravates its overexpression and its

consequences (Fig. 7). The most prescribed T2D drug metformin inhibits PKCε (41) and has

been specifically proposed for diabetic patients with NAFLD and hepatocarcinoma (HCC)

(42). Furthermore, metformin reduces HCC incidence in diabetic patients in a dose-dependent


19

manner (43). We found that metformin also efficiently abolished the expression of insulin-

induced PDGFA mRNA (Fig. S8a), protein content (Fig. S8b) and secretion (Fig. S8c). Thus,

a part of the effects of metformin on insulin sensitivity may be mediated by the reduction of

PDGFA overexpression in hepatocytes.

Discussion

GWAS have only identified so far few genes involved in NAFLD (44) and the

contribution of epigenetics to T2D liver dysfunction is still elusive. While we initially

identified 381 differentially methylated regions (DMRs) in the liver from obese patients with

T2D compared to normal glucose obese patients, we only observed one genome-wide

significant DMR (cg14496282 within PDGFA), associated with the increase of PDGFA

expression in cis, in obese patients with T2D after adjusting for possible confounding effects

of liver steatosis and NASH. The cg14496282 might be instrumental in the functional

impairment of the liver in obesity associated T2D. Notably, we found that liver PDGFA

cg14496282 hypomethylation and concomitant rise in liver PDGFA expression were also

associated with systemic insulin resistance in non-diabetic obese patients but not with their

glucose values. Elevated PDGFA expression was also reported in biliary atresia (45), and is a

diagnostic and prognostic biomarker of cholangiocarcinoma that is a liver cancer associated

with severe insulin resistance (but paradoxically not with obesity) (32,46). Thus, PDGFA

seems to be a liver marker of insulin resistance and of chronic hyperinsulinemia. Furthermore,

the genetic data from our analysis of GRS related to insulin resistance, suggest a causative

effect of plasma insulin levels on methylation level, hepatic expression and secretion of this

growth factor. Therefore, the elevated PDGFA expression in human liver from obese subjects

can be directly due to their severe hyperinsulinemia.


20

PDGFA encodes a dimer disulfide-linked polypeptide (PDGF-AA) that plays a crucial

role in organogenesis (21). The activation of the PDGF-AA receptor signaling is involved in

cirrhotic liver regeneration (22) and the chronic elevation of PDGF-AA in mice liver induces

fibrosis (8). High PDGF-AA levels contribute to hepatic fibrogenesis by activating hepatic

stellate cells in mice (23–25). In human, the association of increased PDGFA expression with

liver steatosis and fibrosis observed in our study and in others supports a similar fibrogenic

role in human liver (19,20), in which chronic hyperinsulinemia might be instrumental (47).

Thus, it possible that the PDGF-AA secreted by the liver in response to chronic

hyperinsulinemia contributes, alone and/or in combination with any of the 380 epigenetically

modified genes, to liver steatosis and fibrogenesis in obese people with diabetes.

We also believe that PDGF-AA may contribute to the inhibitory effect of chronic

hyperinsulinemia on hepatocyte insulin signaling via a feedback autocrine loop (Fig. 6f).

PDGF-AA stimulates its own expression via the activation of PKC. This vicious cycle

perpetuates high PDGF-AA level and thereby continuous insulin resistance. Further studies

will be to determine whether hyperinsulinemia per se and/or accumulation of secreted PDGF-

AA initiate the induction of PDGFA. Our data further suggested that the negative effect of

PDGF-AA on insulin signaling is the consequence of the decrease of IRS1 and INSR, and

PKC activation including PKCƟ and PKCε. These two kinases are known to phosphorylate

IRS-1 and the insulin receptor on serine residues, that impairs the association of the insulin

receptor with IRS proteins, leading to the blockade of AKT activation and of the downstream

signaling (35,36).

Our findings may have a major interest for the treatment of T2D and of its hepatic

complications. We showed that metformin, the most-widely prescribed oral insulin sensitizer

agent prevented the PDGF-AA insulin-induced vicious circle. Metformin has been

specifically proposed for diabetic patients with NAFLD and hepatocarcinoma (HCC) (42).


21

Metformin reduces the risk of HCC incidence in diabetic patients in a dose-dependent manner

(43). Thereby metformin may improve liver insulin sensitivity at least in part through PDGF-

AA liver blockade, explaining its long-term effect against HCC. Beside liver insulin

sensitizers, blocking PDGF-AA activity may be a promising alternative to anti-diabetic

therapeutic. The anti-tumor PDGFR inhibitor imatinib demonstrated unexpected (and

unexplained until now) improvement of insulin sensitivity in insulin-resistant rats (48) as well

as a dramatic blood-glucose-lowering effect in diabetic subjects treated for leukemia (49,50).

Our study also suggests that human epigenome analysis, when directly performed in

disease-affected tissues is an efficient tool to make progress in the pathogenesis of common

diseases. Furthermore, it opens avenues in the identification of new drug targets to combat

T2D, and complications linked to insulin resistance, including NAFLD and cancer.


22

Author contributions

PF, SC, LY and AA designed the study. LY, AA, MC, AB and PF drafted and wrote the

manuscript. LY and MC performed statistical analyses. AB, OS and IR performed the

bioinformatics analysis. SL, JM, JD, GL, LR, MK, MT, JB, GQ, EA, SS, PJ, RB, SGC, VP,

AL and MDC performed the experiments. AA, SC, MC, RC, VR, SL, JM, LR, GL, AL, TD,

PM, BS, AS, JA, CP, JH, AB, FP and PF revised the manuscript. All authors have read and

approved the final version of the manuscript.

Acknowledgements

This study was supported by nonprofit organizations and public bodies for funding of

scientific research conducted in France and within the European Union: “Centre National de

la Recherche Scientifique”, “Université de Lille 2”, “Institut Pasteur de Lille”, “Société

Francophone du Diabète”, “Contrat de Plan Etat-Région”, “Agence Nationale de la

Recherche”, ANR-10-LABX-46, ANR EQUIPEX Ligan MP: ANR-10-EQPX-07-01,

European Research Council GEPIDIAB - 294785. We are grateful to Ms Estelle Leborgne for

helping in the illustrations of the manuscript. The present manuscript has been posted in the

preprint ArXiv server.


23

References

1. Stefan N, Häring H-U. The metabolically benign and malignant fatty liver. Diabetes. 2011 Aug;60(8):2011–7.

2. Bonnefond A, Froguel P. Rare and Common Genetic Events in Type 2 Diabetes: What Should Biologists Know? Cell Metab. 2015 Mar;21(3):357–68.

3. Ling C, Groop L. Epigenetics: A Molecular Link Between Environmental Factors and Type 2 Diabetes. Diabetes. 2009 Dec;58(12):2718–25.

4. Kirchner H, Sinha I, Gao H, Ruby MA, Schönke M, Lindvall JM, et al. Altered DNA methylation of glycolytic and lipogenic genes in liver from obese and type 2 diabetic patients. Mol. Metab. 2016 ;5:171-183

5. Muka T, Nano J, Voortman T, Braun KVE, Ligthart S, Stranges S, et al. The role of global and regional DNA methylation and histone modifications in glycemic traits and type 2 diabetes: A systematic review. Nutr Metab Cardiovasc Dis NMCD. 2016 Jul;26(7):553–66.

6. Nilsson E, Matte A, Perfilyev A, de Mello VD, Käkelä P, Pihlajamäki J, et al. Epigenetic alterations in human liver from subjects with type 2 diabetes in parallel with reduced folate levels. J Clin Endocrinol Metab. 2015 Sep 29;jc.2015-3204.

7. Nilsson E, Jansson PA, Perfilyev A, Volkov P, Pedersen M, Svensson MK, et al. Altered DNA Methylation and Differential Expression of Genes Influencing Metabolism and Inflammation in Adipose Tissue From Subjects With Type 2 Diabetes. Diabetes. 2014 Sep 1;63(9):2962–76.

8. Wahl S, Drong A, Lehne B, Loh M, Scott WR, Kunze S, et al. Epigenome-wide association study of body mass index, and the adverse outcomes of adiposity. Nature. 2017 Jan 5;541(7635):81–6.

9. Thieringer F, Maass T, Czochra P, Klopcic B, Conrad I, Friebe D, et al. Spontaneous hepatic fibrosis in transgenic mice overexpressing PDGF-A. Gene. 2008 Oct 15;423(1):23–8.

10. Caiazzo R, Lassailly G, Leteurtre E, Baud G, Verkindt H, Raverdy V, et al. Roux-en-Y Gastric Bypass Versus Adjustable Gastric Banding to Reduce Nonalcoholic Fatty Liver Disease: A 5-Year Controlled Longitudinal Study. Ann Surg. 2014 Nov;260(5):893.

11. Ahrens M, Ammerpohl O, von Schönfels W, Kolarova J, Bens S, Itzel T, et al. DNA Methylation Analysis in Nonalcoholic Fatty Liver Disease Suggests Distinct Disease-Specific and Remodeling Signatures after Bariatric Surgery. Cell Metab. 2013 Aug 6;18(2):296–302.

12. Bibikova M, Barnes B, Tsan C, Ho V, Klotzle B, Le JM, et al. High density DNA methylation array with single CpG site resolution. Genomics. 2011 Oct 1;98(4):288–95.

13. Voight BF, Kang HM, Ding J, Palmer CD, Sidore C, Chines PS, et al. The Metabochip, a Custom Genotyping Array for Genetic Studies of Metabolic, Cardiovascular, and Anthropometric Traits. PLoS Genet. 2012 Aug 2; 8:e1002793.


24

14. Aryee MJ, Jaffe AE, Corrada-Bravo H, Ladd-Acosta C, Feinberg AP, Hansen KD, et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics. 2014 May 15;30(10):1363–9.

15. Teschendorff AE, Marabita F, Lechner M, Bartlett T, Tegner J, Gomez-Cabrero D, et al. A beta-mixture quantile normalization method for correcting probe design bias in Illumina Infinium 450 k DNA methylation data. Bioinformatics. 2013 Jan 15;29(2):189–96.

16. Samanez CH, Caron S, Briand O, Dehondt H, Duplan I, Kuipers F, et al. The human hepatocyte cell lines IHH and HepaRG: models to study glucose, lipid and lipoprotein metabolism. Arch Physiol Biochem. 2012 Jul;118(3):102–11.

17. Pichard L, Raulet E, Fabre G, Ferrini JB, Ourlin J-C, Maurel P. Human hepatocyte culture. Methods Mol Biol Clifton NJ. 2006;320:283–93.

18. Houseman EA, Kile ML, Christiani DC, Ince TA, Kelsey KT, Marsit CJ. Reference-free deconvolution of DNA methylation data and mediation by cell composition effects. BMC Bioinformatics. 2016;17:259.

19. Murphy SK, Yang H, Moylan CA, Pang H, Dellinger A, Abdelmalek MF, et al. Relationship Between Methylome and Transcriptome in Patients With Nonalcoholic Fatty Liver Disease. Gastroenterology. 2013 Nov 1;145(5):1076–87.

20. Zeybel M, Hardy T, Robinson SM, Fox C, Anstee QM, Ness T, et al. Differential DNA methylation of genes involved in fibrosis progression in non-alcoholic fatty liver disease and alcoholic liver disease. Clin Epigenetics [Internet]. 2015 Mar 14;7(1).

21. Andrae J, Gallini R, Betsholtz C. Role of platelet-derived growth factors in physiology and medicine. Genes Dev. 2008 May 15;22(10):1276–312.

22. Awuah PK, Nejak-Bowen KN, Monga SPS. Role and Regulation of PDGFRα Signaling in Liver Development and Regeneration. Am J Pathol. 2013 May;182(5):1648–58.

23. Hayes BJ, Riehle KJ, Shimizu-Albergine M, Bauer RL, Hudkins KL, Johansson F, et al. Activation of Platelet-Derived Growth Factor Receptor Alpha Contributes to Liver Fibrosis. PLoS ONE [Internet]. 2014 Mar 25;9 :e0092925.

24. Liu X, Brenner DA. Liver: DNA methylation controls liver fibrogenesis. Nat Rev Gastroenterol Hepatol. 2016 Mar;13(3):126.

25. Kocabayoglu P, Lade A, Lee YA, Dragomir A-C, Sun X, Fiel MI, et al. β-PDGF receptor expressed by hepatic stellate cells regulates fibrosis in murine liver injury, but not carcinogenesis. J Hepatol. 2015 Jul 1;63(1):141–7.

26. Scott RA, Lagou V, Welch RP, Wheeler E, Montasser ME, Luan J, et al. Large-scale association analyses identify new loci influencing glycemic traits and provide insight into the underlying biological pathways. Nat Genet. 2012 Sep;44(9):991–1005.

27. Baumeier C, Saussenthaler S, Kammel A, Jähnert M, Schlüter L, Hesse D, et al. Hepatic DPP4 DNA Methylation Associates With Fatty Liver. Diabetes. 2017 Jan 1;66(1):25–35.


25

28. Wong NC, Ng J, Hall NE, Lunke S, Salmanidis M, Brumatti G, et al. Exploring the utility of human DNA methylation arrays for profiling mouse genomic DNA. Genomics. 2013 Jul;102(1):38–46.

29. Shimomura I, Matsuda M, Hammer RE, Bashmakov Y, Brown MS, Goldstein JL. Decreased IRS-2 and Increased SREBP-1c Lead to Mixed Insulin Resistance and Sensitivity in Livers of Lipodystrophic and ob/ob Mice. Mol Cell. 2000 Jul 1;6(1):77–86.

30. Gross SM, Rotwein P. Mapping growth-factor-modulated Akt signaling dynamics. J Cell Sci. 2016 May 15;129(10):2052–63.

31. Mackenzie RW, Elliott BT. Akt/PKB activation and insulin signaling: a novel insulin signaling pathway in the treatment of type 2 diabetes. Diabetes Metab Syndr Obes Targets Ther. 2014 Feb 13;7:55–64.

32. Michelini E, Lonardo A, Ballestri S, Costantini M, Caporali C, Bonati ME, et al. Is cholangiocarcinoma another complication of insulin resistance: a report of three cases. Metab Syndr Relat Disord. 2007 Jun;5(2):194–202.

33. Hilhorst R, Houkes L, Mommersteeg M, Musch J, Berg A van den, Ruijtenbeek R. Peptide Microarrays for Profiling of Serine/Threonine Kinase Activity of Recombinant Kinases and Lysates of Cells and Tissue Samples. In: Gene Regulation [Internet]. Humana Press, Totowa, NJ; 2013 Dec 4; 259–271.

34. Dasgupta S, Bhattacharya S, Maitra S, Pal D, Majumdar SS, Datta A, et al. Mechanism of lipid induced insulin resistance: Activated PKCε is a key regulator. Biochim Biophys Acta BBA - Mol Basis Dis. 2011 Apr;1812(4):495–506.

35. Kim JK, Fillmore JJ, Sunshine MJ, Albrecht B, Higashimori T, Kim D-W, et al. PKC-θ knockout mice are protected from fat-induced insulin resistance. J Clin Invest. 2004 Sep 15;114(6):823–7.

36. Samuel VT, Liu Z-X, Wang A, Beddow SA, Geisler JG, Kahn M, et al. Inhibition of protein kinase Cε prevents hepatic insulin resistance in nonalcoholic fatty liver disease. J Clin Invest. 2007 Mar 1;117(3):739–45.

37. Cenni V, Döppler H, Sonnenburg ED, Maraldi N, Newton AC, Toker A. Regulation of novel protein kinase C epsilon by phosphorylation. Biochem J. 2002 May 1;363(Pt 3):537–45.

38. Liu Y, Graham C, Li A, Fisher RJ, Shaw S. Phosphorylation of the protein kinase C-theta activation loop and hydrophobic motif regulates its kinase activity, but only activation loop phosphorylation is critical to in vivo nuclear-factor-kappaB induction. Biochem J. 2002 Jan 15;361(Pt 2):255–65.

39. Evenou J-P, Wagner J, Zenke G, Brinkmann V, Wagner K, Kovarik J, et al. The Potent Protein Kinase C-Selective Inhibitor AEB071 (Sotrastaurin) Represents a New Class of Immunosuppressive Agents Affecting Early T-Cell Activation. J Pharmacol Exp Ther. 2009 Sep 1;330(3):792–801.


26

40. Marra F, Ghosh Choudhury G, Pinzani M, Abboud HE. Regulation of platelet-derived growth factor secretion and gene expression in human liver fat-storing cells. Gastroenterology. 1994 Oct 1;107(4):1110–7.

41. Rodríguez-Lirio A, Pérez-Yarza G, Fernández-Suárez MR, Alonso-Tejerina E, Boyano MD, Asumendi A. Metformin Induces Cell Cycle Arrest and Apoptosis in Drug-Resistant Leukemia Cells. Leuk Res Treat [Internet]. 2015: e516460

42. Dyson JK, Anstee QM, McPherson S. Republished: Non-alcoholic fatty liver disease: a practical approach to treatment. Postgrad Med J. 2015 Feb;91(1072):92–101.

43. Bo S, Benso A, Durazzo M, Ghigo E. Does use of metformin protect against cancer in Type 2 diabetes mellitus? J Endocrinol Invest. 2012 Feb 1;35(2):231–5.

44. Anstee QM, Day CP. The genetics of NAFLD. Nat Rev Gastroenterol Hepatol. 2013 Nov;10(11):645–55.

45. Cofer ZC, Cui S, EauClaire SF, Kim C, Tobias JW, Hakonarson H, et al. Methylation Microarray Studies Highlight PDGFA Expression as a Factor in Biliary Atresia. PLOS ONE. 2016 Mar 24;11(3):e0151521.

46. Boonjaraspinyo S, Boonmars T, Wu Z, Loilome W, Sithithaworn P, Nagano I, et al. Platelet-derived growth factor may be a potential diagnostic and prognostic marker for cholangiocarcinoma. Tumour Biol J Int Soc Oncodevelopmental Biol Med. 2012 Oct;33(5):1785–802.

47. Bril F, Lomonaco R, Orsak B, Ortiz-Lopez C, Webb A, Tio F, et al. Relationship between disease severity, hyperinsulinemia, and impaired insulin clearance in patients with nonalcoholic steatohepatitis. Hepatology. 2014 Jun 1;59(6):2178–87.

48. Hägerkvist R, Jansson L, Welsh N. Imatinib mesylate improves insulin sensitivity and glucose disposal rates in rats fed a high-fat diet. Clin Sci. 2008 Jan 1;114(1):65–71.

49. Breccia M, Muscaritoli M, Aversa Z, Mandelli F, Alimena G. Imatinib Mesylate May Improve Fasting Blood Glucose in Diabetic Ph+ Chronic Myelogenous Leukemia Patients Responsive to Treatment. J Clin Oncol. 2004 Nov 15;22(22):4653–5.

50. Veneri D, Franchini M, Bonora E. Imatinib and Regression of Type 2 Diabetes. N Engl J Med. 2005 Mar 10;352(10):1049–50.


27

Legends of Figures

Fig. 1. a) Quantile-quantile (qq-) plot showing the residual inflation of test statistics before

and after genomic-control correction. b) Manhattan plot centered on PDGFA cg14496282

methylation site showing association signal within PDGFA bounds. Hepatic Pdgfa expression

in c) 6 weeks old male B6 mice that were diet-induced obese (DIO) responder (Resp, black

circle, n= 12) and DIO-non-responder (nResp, white circle, n=10) and in d) BXD mice (n=

45) fed on Chow diet (CD, green circle) or a HFD (black circle) fed for 21 weeks. ***

indicates p value < 0.0001 by unpaired t test with Welch’s correction.

Fig. 2. a) PDGF-AA secretion from IHH cells and primary human hepatocytes was measured

by ELISA kit. b) Measurement of insulin-induced AKT phosphorylation in response to

human insulin (NovoNordisk) for the indicated times. IHH cells were incubated in a culture

medium containing 5 mM Glucose, 2 % FCS with or without 100 nM human insulin for the

indicated times. AKT phosphorylation was stimulated by 200 nM insulin for one hour.

Immunoblotting for phospho-AKT (p-AKT) was done using the anti phospho-AKT (Serine

473) antibodies. The Fig. shows the result of a representative experiment out of three. c)

Effect of insulin on the glycogen production. Insulin-induced glycogen production was

measured by ELISA in IHH cells that were pre-cultured with 100 nM insulin for 16 h and 24

h. d) Increase of PDGFA mRNA by insulin. IHH cells were cultured with 100 nM human

insulin for 16 h and 24 h. The PDGFA mRNA level was quantified by qRT-PCR and

normalized against RPLP0. The expression levels from untreated cells were set to 100 %.

Data are the mean ± SEM (**: p < 0.001). e) PDGF-AA abundance in IHH cells cultured with

insulin. IHH cells were cultured with 100 nM human insulin for the indicated times. PDGF-

AA content was quantified by Western Blotting experiments. The blot is one representative


28

out of three independent experiments. f) PDGF-AA secretion in response to insulin. IHH cells

were cultured with insulin for the indicated times. The measurement of PDGF-AA by ELISA

from the supernatant that was retrieved of IHH cells cultured . g) Methylation levels at

PDGFA cg14496282 in response to insulin. IHH cells were cultured in a culture medium

containing 5 mM Glucose, 2 % FCS with or without 100 nM human insulin for 24 h.

Methylation level at the cg14496282 was quantified by the Infinium HumanMethylation450

BeadChip. h) Effect of chronic insulin in insulin-induced AKT activation in HepG2 cells.

HepG2 cells were incubated in a culture medium with 100 nM human insulin or without (Ctl)

for 24 h. AKT phosphorylation was stimulated by 200 nM insulin for one hour.

Immunoblotting for phospho-AKT (p-AKT) was done using the anti phospho-AKT (Serine

473) antibodies. The Fig. shows the result of a representative experiment out of three. i)

Increase of PDGFA mRNA by insulin in HepG2 cells. The PDGFA mRNA level was

quantified by qRT-PCR in HepG2 and IHH cells that were cultured with insulin for 24 h. The

PDGFA mRNA was normalized against RPLP0. The expression levels from untreated cells

were set to 100 %. Data are the mean ± SEM (**: p < 0.001).

Fig. 3. Effects of a) human PDGF-AA recombinant, b) PDGFA blocking antibodies or c) the

PDGFR inhibitor ki11502 on insulin-induced AKT activation. Activation of AKT was

monitored by western blotting using total proteins from IHH cells that were cultured with the

human recombinant PDGF-AA at the indicated concentrations for 24 h, which subsequently

were incubated with 200 nM insulin for stimulating AKT phosphorylation. For a-c, IHH cells

were co-incubated in a culture medium containing 5 mM Glucose, 2 % FCS with or without

100 nM human insulin for 24 hours plus a) PDGF-AA at the indicated concentration, b)

PDGFA antibodies (+; 0.75 µg or ++; 1.5 µg) or c) ki11502 at the indicated concentration.

The Figures show the result of a representative experiment out of three. d) Effect of the


29

PDGFR inhibitor ki11502 on the glycogen production. Glycogen was measured by ELISA in

IHH cells that were co-cultured with 5 µM ki11502 and insulin for the indicated times.

Glycogen was monitored after stimulating cells in a KRP buffer without (Ctl) or with insulin

for 1 h and 20 mM glucose. Glycogen was monitored by ELISA.

Fig. 4: a) Volcano plot showing differences in putative serine/threonine kinase activities

between control and insulin-treated IHH cells for 24 h. Specific and positive kinase statistic

(in red) show higher activity in IHH cultured with insulin compared with control samples.

Effects of b) insulin and c) PDGF-AA on the phosphorylation of PKCƟ and PKCε. IHH cells

were cultured with insulin for the indicated times or PDGF-AA (for 24 h). Phosphorylation of

PKCƟ (Ser 676) and PKCε (Ser 729) were measured by western blotting and normalized

against total PKCƟ and PKCε. d) Effect of insulin and PDGF-AA on the tyrosine

phosphorylation (Y972) of INSR (p-IR) and threonine phosphorylation of INSR (T1376),

INSR abundance (IR) and αTubulin. Western Blotting experiments were achieved from total

proteins of IHH cells cultured either with 100 nM insulin of 100 ng/ml PDGF-AA for 24 h.

Phosphorylation of INSR was done by stimulating IHH cells with insulin for 1 h. The Figures

show the result of a representative experiment out of three.

Fig. 5: Effect of insulin on a) IRS1 mRNA level and b) protein. The IRS1 mRNA level and

IRS1 abundance was quantified by qRT-PCR and Western Blotting in IHH cells cultured with

100 nM insulin for 24 h. The IRS1 mRNA was normalized against RPLP0. The expression

levels from untreated cells were set to 100 %. Data are the mean ± SEM (**: p < 0.001). c)

effect of PDGF-AA on the IRS1 content. IHH cells were cultured with PDGF-AA at the

indicated concentrations for 24 h. The Figures show the result of a representative experiment


30

out of three. d) Effect of siRNA against IRS1 on the insulin-induced AKT activation.

Duplexes of small interfering RNAs were transfected in IHH cells for 48 h. Thereafter, AKT

phosphorylation on the serine 473 was induced with insulin for 1 h. The Figures show the

result of a representative experiment out of three. Effect of e) and f) the PKC inhibitor

sotraustorin PKC and f) and g) phorbol 12-myristate 13-acetate (PMA) PKC activator on the

expression of IRS1 mRNA and protein. PDGFA mRNA was quantified in IHH cells cultured

with either sotrastaurin at the indicated concentration in the presence of 100 nM insulin for 24

h, or PMA for the indicated times. The IRS1 mRNA was normalized against RPLP0. The

expression levels from untreated cells were set to 100 %. Data are the mean ± SEM of three

independent experiments made in triplicates (***: p < 0.0001).

Fig. 6: Effect of ki11502 PDGFR inhibitor on the PDGFA expression in response to a)

PDGF-AA and b) insulin. IHH cells were cultured for 24 h with 100 ng/ml PDGF-AA or 100

nM insulin in the presence or absence of 5 µM ki11502 for 24 h. Effect of c) PKC activator

phorbol 12-myristate 13-acetate (PMA) and d-e) the PKC inhibitor sotraustorin on the

expression of PDGFA mRNA induced by d) insulin or e) PDGF-AA .The PDGFA mRNA

was quantified by qRT-PCR in IHH cells cultured with either PMA for the indicated times, or

100 nM insulin or 100 ng/ml PDGF-AA in the presence of 1 µM sotrastaurin or at the

indicated concentration for 24 h. The PDGFA mRNA was normalized against RPLP0. The

expression levels from untreated cells were set to 100 %. Data are the mean ± SEM of three

independent experiments made in triplicates (*, p<0.05; ***: p < 0.0001).

Fig. 7: Schematic representation of the mechanism linking hyperinsulinemia to hepatic insulin

resistance in T2D. Insulin promotes hypomethylation and the rise of PDGFA expression,


31

leading to PDGF-AA secretion. In turn, PDGF-AA inhibits the insulin signaling, in a negative

autocrine feedback loop, via a mechanism involving a decrease in the IRS1 and INSR

abundance and PKC (PKCƟ and PKCε) activation.

Tables

Table 1. Association of liver methylation levels of cg14496282 and liver PDGFA gene expression with multiple

quantitative and binary traits. Methylation levels at cg14496282 and PDGFA gene expression are the

endogenous variable in all linear regressions used to measure associations. SD: Standard Deviation.

Traits(unit)

PDGFA cg14496282 methylation PDGFA

Expression

Effect size in % of methylation / trait unit

(p-value)

Effect size in SD/trait unit

(p-value)

Controls T2D cases Controls T2D cases

cg14496282 methylation (%)

-1.44

(6.27×10-3

) -2.497

(4.94×10-3

)

PDGFA expression (Scaled -SD)

-0.0548 (6.27×10

-3)

-0.0338 (4.94×10

-3)

Fasting glucose (mmol/l)

-0.0112 (0.79)

- 0.292 (0.196)

-

Fasting insulin (pmol/l)

-1.45×10-3

(2.32×10

-3)

- 6.83×10

-3

(9.49×10-3

) -

HOMA2-B (unitless - log)

-0.169 (2.92×10

-3)

- 0.626

(0.038) -

HOMA2-IR (unitless - log)

-0.104 (4.93×10

-3)

- 0.528

(7.47×10-3

) -

QUICKI (unitless)

1.66 (0.01)

- -9.192

(9.78×10-3

) -

Steatosis (%) -2.15×10

-3

(0.01) -4.34×10

-4

(0.42) 0.0136

(2.72×10-3

) 0.020

(2.14×10-6

)

NASH (Yes/ No) -0.17 (0.04)

-0.072 (0.03)

2.115 (9.38×10

-7)

1.447 (3.37×10

-8)

Hepatic fibrosis (Yes/ No)

-0.07 (0.09)

-0.051 (0.04)

0.187 (0.434)

0.631 (2.66×10

-3)

Alanine aminotransferase (UI/L)

-1.44×10-4

(0.89) -1.34×10

-3

(0.03) 0.0106 (0.067)

0.0194 (1.46×10

-4)

Aspartate aminotransferase (UI/L)

-4.72×10-3

(0.06) -1.76×10

-3

(0.04) 0.0342

(7.89×10-3

) 0.0327

(2.56×10-6

)


254x338mm (300 x 300 DPI)


190x254mm (144 x 144 DPI)


190x254mm (96 x 96 DPI)


of 66Diabetes

190x254mm (96 x 96 DPI)


Controls (n=96) T2D cases (n=96)

Traits (unit) Mean (SD) or n (%) Mean (SD) or n (%) p-value

cg14496282 methylation (%) 60.3% (17.2%) 41.4% (12.2%) 9.27×10-16

PDGFA expression (log10-SD) 2.79 (0.172) 2.91 (0.165) 1.70×10-6

Age (years) 46.8

(7.01)

48.2

(6.34)

0.13

BMI (kg/m²) 47.1

(7.4)

49.1

(7.5)

0.06

Fasting glucose (mmol/l) 5.2

(0.407)

8.5

(3.2)

2.2×10-19

Fasting insulin (pmol/l) 78.8

(36.497)

190.5

(319)

0.029

HOMA2-B (unitless) 113

(34.8)

71.5

(42.4)

4.67×10-11

HOMA2-IR (unitless) 1.47

(0.672)

2.14

(1.435)

8.0×10-5

QUICKI (unitless) 0.34

(0.026)

0.313

(0.036)

3.83×10-7

Steatosis (%) 21.3%

(20.1%)

42.4%

(23.8%)

4.77×10-10

NASH* (Yes/ No) 4

(4.2%)

17

(18%)

4.44×10-3

Hepatic fibrosis* (Yes/ No) 21

(21%)

45

(45%)

6.55×10-4

Alanine aminotransferase (UI/L)

27.1

(15.8)

34.3

(19.8)

6.04×10-3

Aspartate aminotransferase (UI/L)

22.344

(7.04)

26.9

(14.3)

5.5×10-3

Table S1. Clinical characteristics of the 192 samples (96 cases + 96 controls) included in the discovery cohort. Quantitative traits were compared

between cases and controls using unadjusted linear regression and binary traits* were compared using Fisher exact test.


Table S2. Clinical characteristics of the 65 selected samples (12 cases + 53 controls) used for replication. Quantitative traits were

compared between cases and controls using unadjusted linear regression and binary traits* were compared using Fisher exact

test.

Controls (n=53) T2D cases (n=12)

Traits (unit) Mean (SD) or n (%) Mean (SD) or n (%) p-value

cg14496282 methylation (%) 63.3 (19.2) 41.4 (11.6) 3.59e-04

Age (years) 46.7 (13) 54.6 (13.1) 0.063

Sex (female)* 42 (79.2) 5 (41.7) 0.014

BMI (kg/m2) 40.6 (13.6) 43.2 (11.5) 0.542


SNP

Closest gene (locus) Fasting insulin raising allele Effect size

(p-value)

in Controls

Effect size (p-value)

in T2D cases

I2

statistic

(p-value)

Fixed effect meta-analysis

effect size

(p-value)

rs459193

ANKRD55

MAP3K1

G

-1.85

( 0.483 )

-0.87

( 0.691 )

0%

( 0.776 )

-1.27% ( 0.448 )

rs4865796 ARL15 A

-3.53

( 0.216 )

0.96

( 0.656 )

37.49%

( 0.206 )

-0.68% ( 0.692 )

rs3822072 FAM13A1 A

-2.89

( 0.267 )

0.12

( 0.953 )

0%

( 0.353 )

-0.97% ( 0.532 )

rs1421085 FTO C

-0.66

( 0.794 )

-2.83

( 0.124 )

0%

( 0.486 )

-2.09% ( 0.158 )

rs10195252

GRB14

COBLL1

T

-0.27

( 0.913 )

0.22

( 0.91 )

0%

( 0.876 )

0.03% ( 0.983 )

rs1167800 HIP1 A

2.31

( 0.414 )

0.54

( 0.793 )

0%

( 0.611 )

1.15% ( 0.486 )

rs2943645 IRS1 T

-1.66

( 0.552 )

-0.73

( 0.698 )

0%

( 0.783 )

-1.02% ( 0.511 )

rs4846565 LYPLAL1 G

-6.69

( 0.006 )

1.49

( 0.483 )

85.03%

( 0.01 )

-2.15% ( 0.171 )

rs6822892 PDGFC A

-0.88

( 0.731 )

-1.03

( 0.62 )

0%

( 0.963 )

-0.97% ( 0.546 )

rs731839 PEPD G

-3.69

( 0.197 )

-0.95

( 0.643 )

0% ( 0.432 ) -1.88% ( 0.256 )

rs17036328 PPARG T

-6.3

( 0.122 )

-4.42

( 0.115 )

0%

( 0.701 )

-5.02% ( 0.028 )

rs2126259 PPP1R3B T

-0.24

( 0.96 )

-2.54

( 0.418 )

0%

( 0.69 )

-1.86% ( 0.477 )

rs2745353 RSPO3 T

-4.72

( 0.056 )

-1.23

( 0.546 )

17.56%

( 0.271 )

-2.66% ( 0.088 )

rs7903146 TCF7L2 C

-1.66

( 0.619 )

-2.75

( 0.172 )

0%

( 0.78 )

-2.46% ( 0.15 )

rs974801 TET2 G

4.32

( 0.141 )

-2.37

( 0.229 )

72.58%

( 0.056 )

-0.29% ( 0.861 )

rs6912327 UHRF1BP1 T

-1.68

( 0.576 )

-3.87

( 0.064 )

0%

( 0.546 )

-3.16% ( 0.062 )

rs1530559 YSK4 A

3.94

( 0.154 )

-1.71

( 0.356 )

65.92%

( 0.087 )

0.05% ( 0.974 )

rs860598 IGF1 A

-5.39

( 0.118 )

3.6

( 0.095 )

79.95%

( 0.026 )

1.09% ( 0.548 )

rs780094 GCKR C

0.21

( 0.936 )

1.35

( 0.513 )

0%

( 0.736 )

0.92% ( 0.569 )

Overall

Meta-analysis

(Cases +

Controls)

- -

-1.63%

(0.011)

-0.83%

(0.079)

2.4%

(0.428)

-1.11%

(0.004)

Genetic Risk

Score (GRS)- -

-1.

(0.010) (0.099) 30.2% (0.231)

-1.05%

(0.004)

Table S3. Association between fasting insulin raising alleles and DNA methylation at cg14496282. Associations were assessed using linear regression with

methylation as response variable. All models were adjusted for age, BMI, total cholesterol, HDL cholesterol, triglycerides and fasting glucose. Effect sizes are

reported as percentage of DNA methylation per allele.


Gene Log2 Fold Change p-value

PDGFRB -0.68 3.2E-10

IL1RN -0.59 3.0E-15

A2M -0.33 2.2E-09

FGF2 -0.31 9.5E-03

VDR -0.25 3.2E-02

PRKACB -0.25 1.7E-03

IRS1 -0.21 1.1E-02

POU5F1 -0.19 6.0E-03

NFATC2 -0.17 2.7E-02

NR3C1 -0.14 4.4E-02

PRKACA -0.13 2.1E-02

PDPK1 -0.13 1.8E-02

HDAC1 -0.12 3.9E-02

RXRB -0.12 4.2E-02

PRKAR1A -0.12 2.4E-02

SP1 -0.11 3.9E-02

NME2 0.11 2.6E-02

SSRP1 0.12 2.2E-02

SPP1 0.12 1.6E-02

NCL 0.12 8.2E-03

TAF4 0.16 8.2E-03

NFKB1 0.17 2.8E-03

FURIN 0.18 4.5E-02

VEGFB 0.20 1.5E-02

PDAP1 0.21 3.0E-03

HIF1A 0.22 6.4E-05

TGFB1 0.23 3.0E-04

MAP2K1 0.23 1.3E-04

MAP2K2 0.23 1.1E-02

DNM2 0.25 1.7E-04

RRAS 0.25 4.4E-02

TNFRSF1A 0.25 7.6E-06

GRB14 0.26 2.2E-02

NFIC 0.27 1.3E-02

NME1 0.28 1.1E-06

IL18 0.29 3.1E-02

PDGFB 0.43 1.2E-04

PCSK5 0.52 8.3E-12

EDN1 0.55 6.4E-05

VEGFA 0.56 5.8E-13

NES 0.64 1.2E-25

DUSP1 0.67 9.7E-11

PDGFA 0.80 1.1E-11

KLF5 0.87 3.6E-13

EGR1 1.13 2.2E-16

Table S4. List of deregulated genes within the network of PDGFA


Gene Log2 Fold

Change p-value Gene

Log2 Fold

Change p-value Gene

Log2 Fold

Change p-value

G6PC -1,69 2,3E-36 FAS -0,14 2,3E-02 TCF7L2 0,24 5,7E-04

PDGFRB -0,68 3,2E-10 CPT1A -0,14 1,9E-02 OAS1 0,25 2,3E-03

IL1RN -0,59 3,0E-15 ALG2 -0,14 1,2E-02 SREBF1 0,25 8,4E-03

GPAM -0,54 1,5E-18 ACADM -0,14 2,2E-02 TIGAR 0,25 5,1E-04

SLC2A2 -0,53 3,8E-14 PIGF -0,14 3,5E-02 HEXB 0,25 1,1E-05

CREB3L3 -0,51 1,8E-05 CHKB -0,13 4,7E-02 ITGB1 0,25 6,3E-06

ENPP2 -0,47 2,5E-04 GM2A -0,13 3,2E-02 TNFRSF1A 0,25 7,6E-06

GNMT -0,47 2,9E-04 PDPK1 -0,13 1,8E-02 CTGF 0,25 7,8E-03

PPARGC1A -0,46 1,3E-08 EXTL2 -0,12 4,2E-02 CHST6 0,26 2,9E-02

ETNK2 -0,44 1,3E-09 PHKB -0,12 3,5E-02 UGP2 0,26 3,0E-04

SLC23A2 -0,41 2,2E-14 NAGA -0,12 3,9E-02 GCNT3 0,26 5,4E-04

SH3YL1 -0,39 6,4E-07 MAN2A1 -0,11 4,8E-02 B3GNT3 0,27 8,0E-05

PXYLP1 -0,38 4,9E-04 B3GAT1 -0,11 3,5E-02 GYS1 0,27 4,5E-05

CYP3A5 -0,35 6,1E-03 PGM3 -0,10 4,3E-02 ADRBK1 0,28 2,0E-04

TRPV1 -0,35 2,0E-05 NQO1 0,10 4,6E-02 CSF1 0,29 3,5E-03

PGAP1 -0,34 1,7E-05 RHOA 0,10 3,8E-02 IL18 0,29 3,1E-02

PMM1 -0,34 3,0E-06 CLTC 0,10 4,5E-02 GPI 0,29 2,4E-08

NEU3 -0,33 7,4E-05 XBP1 0,10 2,9E-02 NR1D1 0,30 3,1E-03

PPP1R3F -0,31 9,2E-03 SCD 0,11 3,6E-02 GAL 0,30 2,7E-02

ST6GAL1 -0,31 3,2E-10 SPP1 0,12 1,6E-02 B4GALNT1 0,30 5,2E-04

FGF2 -0,31 9,5E-03 PTGFRN 0,12 4,5E-02 SFN 0,31 1,5E-03

GALNT7 -0,29 3,6E-02 GNB1 0,12 5,2E-03 TPI1 0,31 5,5E-07

RGN -0,29 1,6E-05 PTEN 0,13 3,5E-02 HBEGF 0,31 2,1E-02

PIGV -0,27 9,9E-05 SH3KBP1 0,13 2,9E-02 IGF2 0,32 8,2E-05

SLC37A4 -0,27 1,2E-07 DSE 0,13 4,8E-02 GOT1 0,33 4,0E-07

PPARA -0,27 4,3E-05 GALE 0,13 1,8E-02 SLC2A8 0,33 2,3E-04

GALT -0,25 9,8E-05 AFF4 0,13 4,2E-02 PPARG 0,33 2,7E-04

IGF1R -0,25 1,8E-05 PRPS1 0,13 3,0E-02 CHSY1 0,34 6,5E-08

NDST2 -0,25 1,2E-04 CDC42 0,14 1,5E-02 GMDS 0,36 1,5E-09

C5 -0,25 2,0E-04 SLC9A3R1 0,14 3,1E-02 PPP1R3C 0,37 1,8E-04

GCKR -0,25 1,0E-03 TALDO1 0,14 7,0E-03 NTSR1 0,37 4,9E-03

ONECUT1 -0,25 3,3E-02 ARF6 0,15 7,0E-03 GBE1 0,38 8,8E-06

LIPC -0,25 5,7E-03 RPE 0,15 1,7E-02 DKK1 0,39 8,5E-08

GPLD1 -0,24 2,4E-03 GNPDA1 0,15 1,3E-02 SERINC2 0,39 3,3E-05

ENPP1 -0,24 3,9E-05 GK 0,15 4,2E-02 FOXA2 0,39 2,8E-07

NDST1 -0,24 7,4E-04 GLA 0,15 1,3E-02 PKM 0,40 1,6E-12

CTBS -0,22 9,4E-03 SH3GLB1 0,16 1,1E-02 PLCH2 0,40 3,5E-03

H6PD -0,22 8,5E-04 PYGB 0,16 3,0E-02 PGK1 0,42 5,4E-07

IRS1 -0,21 1,1E-02 HS2ST1 0,17 4,7E-03 PLAU 0,43 9,5E-04

ALDH5A1 -0,21 2,8E-04 GNE 0,17 6,1E-03 PDGFB 0,43 1,2E-04

S1PR2 -0,21 3,0E-02 CEBPB 0,17 3,6E-02 PLA2G3 0,44 5,4E-04

KHK -0,21 3,6E-03 XYLT1 0,17 3,2E-02 PPP1R3G 0,45 1,6E-04

HDAC5 -0,20 3,4E-03 COQ2 0,17 2,2E-02 STBD1 0,45 2,6E-06

PPP1R3E -0,20 2,2E-03 NFKB1 0,17 2,8E-03 ALDOA 0,45 6,8E-11

SIAE -0,20 7,8E-04 FOXO1 0,18 2,4E-02 ICAM1 0,45 1,2E-17

PLCD1 -0,20 4,9E-02 G6PD 0,18 3,0E-02 ALDOC 0,46 1,3E-10

CHPT1 -0,19 6,3E-04 FUCA2 0,18 8,3E-04 DUSP6 0,46 2,0E-15

PIGZ -0,19 2,9E-02 TKT 0,18 8,3E-03 SLC2A1 0,49 1,3E-18

SLC5A2 -0,19 1,8E-02 MTMR2 0,19 3,9E-03 GRB10 0,49 1,8E-06

CHST3 -0,17 1,4E-03 GYG1 0,19 5,5E-03 PFKP 0,52 3,8E-09

EPHX1 -0,17 3,3E-03 INPP5E 0,19 3,9E-02 AGPAT2 0,53 3,1E-06

FUCA1 -0,17 1,3E-02 ADORA2B 0,20 1,2E-02 SLC16A3 0,54 2,7E-10

LCAT -0,17 4,3E-02 MYC 0,20 2,3E-04 EDN1 0,55 6,4E-05

PIGP -0,16 3,2E-02 TRPV2 0,20 3,7E-03 ANGPTL8 0,56 1,6E-06

SULF2 -0,16 3,1E-02 NEU1 0,21 1,5E-05 HK1 0,57 2,1E-05

SERINC5 -0,16 2,6E-02 GAPDH 0,21 1,5E-04 CHST15 0,62 1,6E-10

TGFBR1 -0,16 1,3E-02 PLCG2 0,21 2,7E-02 CEMIP 0,71 1,7E-20

HECTD4 -0,16 4,8E-02 PRKAA2 0,21 2,2E-03 PFKFB3 0,71 1,3E-07

DYRK2 -0,16 2,8E-02 HS6ST1 0,22 7,7E-03 IGFBP3 0,78 1,4E-08

CPS1 -0,15 1,1E-02 HIF1A 0,22 6,4E-05 PPP1R15A 0,80 3,5E-19

PLCB1 -0,15 2,7E-02 PFKFB2 0,22 5,3E-04 PDGFA 0,80 1,1E-11

EPM2AIP1 -0,15 1,2E-02 CHKA 0,22 1,9E-05 HK2 0,82 8,4E-23

SRD5A3 -0,14 3,9E-02 TGFB1 NR4A1 0,86 1,0E-21

PLSCR1 -0,14 2,9E-02

Table S5. List of deregulated genes within the network related to the metabolism of carbohydrates


Table S6. Correlation between multiple genetic risk scores (GRS) and PDGFA methylation in

192 ABOS study participants. GRS are calculated as the number of trait/risk increasing alleles

over NSNP (number of SNPs) independent loci.

Traits Publication identifying genome-

wide associated SNPs

Correlation between

GRS and PDFGA DNA

methylation

at cg14496282

(p-value)

Type 2 diabetes

(NSNP=65) Morris et al.

Nat. Genet. (2012)

-0.07

(0.32)

BMI

(NSNP =97)

Locke et al.

Nature (2015)

-0.01

(0.87)

Fasting glucose

(NSNP =24)

Vaxillaire et al.

Diabetologia (2014)

-0.05

(0.51)

Fasting insulin

(NSNP =19)

Scott et al.

Nat. Genet. (2012)

-0.21

(0.005)


Supplemental Information

Supplemental Experimental Procedures

Microarray mRNA expression analysis. Transcriptome profiling was performed using the

HumanHT-12 v4.0 Whole-Genome DASL HT Assay (Illumina). Total RNA was converted to

cDNA using biotinylated oligo-dT18 and random nonamer primers, followed by

immobilization to a streptavidin-coated solid support. The biotinylated cDNAs were then

simultaneously annealed to a set of assay-specific oligonucleotides based on content derived

from the National Center for Biotechnology Information (NCBI) Reference Sequence

Database (release 98). The extension and ligation of the annealed oligonucleotides generate

PCR templates that are then amplified using fluorescently-labeled (P1) and biotinylated (P2)

universal primers. The labeled PCR products were captured on streptavidin paramagnetic

beads, to yield single-stranded fluorescent molecules which were then hybridized, via gene-

specific complementarity, to the HumanHT-12 BeadChip, thereafter fluorescence intensity

was measured for each bead. Hybridized chips were scanned by using iScan (Illumina) and

raw measurements were extracted by GenomeStudio software version 3.0 (Illumina).

Epigenome-wide DNA methylation profiling. Bisulfite converted DNA was amplified,

fragmented and hybridized to the BeadChips following the standard Infinium protocol. All the

samples were randomized across the chips and analyzed on the same machine by the same

technician to reduce batch effects. After single base extension and staining, the BeadChips

were imaged with the Illumina iScan. Raw fluorescence intensities of the scanned images

were extracted with the GenomeStudio (V2011.1) Methylation module (1.9.0) (Illumina). The

fluorescence intensity ratio was used to calculate the β-value which corresponds to the

methylation score for each analyzed site according to the following equation: β-

value = intensity of the Methylated allele (M) / (intensity of the Unmethylated allele (U) +

intensity of the Methylated allele (M) + 100). DNA methylation β-values range from zero

(completely unmethylated) to one (completely methylated). All samples had high bisulfite

conversion efficiency (signal intensity >4000) and were included for further analysis based on

GenomeStudio quality control steps where control probes for staining, hybridization,

extension and specificity were examined. The intensity of both sample dependent and sample

independent built in controls was checked for the red and green channels using

GenomeStudio.

Intermediate metabolic traits. Body mass index (BMI) is the weight in kilograms divided

by the square of the height in meters. The HOMA2-IR and HOMA2-B indexes were

calculated using the HOMA (HOmeostasis Model Assessment) calculator based on fasting

glucose and fasting insulin levels in subjects in steady-state situation (available at

http://www.dtu.ox.ac.uk/homacalculator). We also calculated the Quantitative Insulin

Sensitivity Check Index (QUICKI) using fasting glucose and fasting insulin levels (available

at https://sasl.unibas.ch/11calculators-QUICKI.php). The degree of steatosis was defined as

the percentage of hepatocytes containing fat droplets. In the control group of normoglycemic

samples, the association levels reported in our study cannot be influenced by T2D treatment.

SNP genotyping, ethnic characterization and genetic risk score. For HapMap, 272

subjects including 87 of European ancestries [HapMap CEU], 97 of Asian ancestries

[HapMap CHB] and 88 of African ancestries [HapMap YRI]) genotype calls at the 106,470


SNPs present on the Metabochip were available. The first two components were sufficient to

discriminate ethnic origin (Fig. S7) and we observed that study participants clustered well

with HapMap samples of European ancestries. We used 19 SNPs previously established for

their association with fasting insulin to assess the possibility of causal direct link between

DNA methylation at cg14496282. These 19 SNPs were on the Metabochip and all passed our

quality control. These SNPs as well as the associated insulin raising alleles are reported in

Table S3. We assessed the combined effect of these SNPs on DNA methylation using either

fixed-effect meta-analysis and by testing the association of DNA methylation at cg14496282

with a genetic risk score (GRS) defined for each individual as the count of the number of

fasting insulin raising alleles. We also tested the association between three other GRS (T2D,

BMI and fasting glucose raising alleles) and DNA methylation at cg14496282 and expression

in Table S6.

DNA and RNA isolation. The DNA samples were isolated using the Gentra Puregene Tissue

Kit (Qiagen, Les Ulis, France). In order to improve isolation, DNA was then resuspended in

Tris/Hcl/EDTA buffer and precipitated by adding chloroform:isoamyl alcohol (4 %). Purity

was determined by using a NanoDrop (NanoDrop Technologies, Wilmington, USA) and

concentration was determined by using the Qubit® dsDNA BR Assay Kit (Life Technologies

a brand of Thermo Fisher Scientific, Saint Aubin, France). The RNeasy® Lipid Tissue kit

(Qiagen) and QIAamp RNA Blood Mini Kit (Qiagen) were used to isolate RNA from liver

tissue and blood, respectively. Quality and integrity were tested on 2100 Bioanalyzer

Instruments by using a RNA 6000 Nanochip (Agilent Technologies, Les Ulis, France).

Concentration was determined by using the Qubit® RNA BR Assay Kit (Life Technologies).

Oil Red O staining. IHH cells were grown in a 24-well plate (at an initial density of 105

cells/well) and incubated with 100 nM insulin for 16 hrs and 24 hrs. Cells were then washed

three times with iced PBS and fixed with 4% paraformaldehyde for 30 minutes. After

fixation, cells were washed three times and stained with Oil Red O solution for 15 min at

room temperature. For Oil Red O content levels quantification, DMSO was added to sample;

after shaking at room temperature for 5 min, the density of samples was read at 490 nm on a

spectrophotometer

Cell proliferation and apoptosis. IHH cells were cultured with 100 nM insulin at different

incubation times. Cell number was counted by trypan blue and apoptosis was determined by

scoring cells displaying pycnotic or fragmented nuclei (visualized with Hoechst 33342). The

counting was performed blind by two different experimenters.

RNA-sequencing. RNA samples with a RNA integrity number higher than 9.0 were extracted

from IHH cells treated or not with 200 nM insulin for 24 hours, from at least three

independent experiments. RNA libraries were prepared using the TruSeq Stranded mRNA

Library Preparation Kit (Illumina, San Diego, CA, USA) following the manufacturer’s

instructions. The libraries were sequenced with the NextSeq 500 (Illumina). A mean of 50

million paired-end reads of 100 bp were generated for each sample. More than 92% of the

reads for each library were effectively mapped to the hg19 human genome assembly using

TopHat2. Subsequently, both quantification and annotation of the reads were performed using

Rsubread. Finally, the differential gene expression analyses (i.e. 24 h insulin-treated IHH cells

versus IHH cells at baseline) were performed using DESeq2. The differentially expressed

genes were subjected to Ingenuity Pathway Analysis (Qiagen, Hilden, Germany) to decipher


the major biological pathways and diseases emphasized by the significantly deregulated genes

(with a p-value < 0.05).

PamChip peptide microarrays for kinome analysis. kinome analysis was achieved using

Serine Threonine Kinases microarrays, which were purchased from PamGene International

BV. Each array contained 140 target peptides as well as 4 control peptides. Sample

incubation, detection, and analysis were performed in a PamStation 12 according to the

manufacturer’s instructions. Briefly, extracts from IHH cells cultured with insulin for 24

hours were made using M-PER mammalian extraction buffer (Thermo Scientific) for 20

minutes on ice. The lysates were then centrifuged at 15,871 g for 20 minutes to remove all

debris. The supernatant was aliquoted, snap-frozen in liquid nitrogen, and stored at –80°C

until further processing. Prior to incubation with the kinase reaction mix, the arrays were

blocked with 2% BSA in water for 30 cycles and washed 3 times with PK assay buffer.

Kinase reactions were performed for 1 hour with 5 µg of total extract for the mouse

experiment or 2.5 µg of total extract for the mature adipocyte and 400 µM ATP at 30°C.

Phosphorylated peptides were detected with an anti-rabbit–FITC antibody that recognizes a

pool of anti–phospho serine/threonine antibodies. The instrument contains a 12-bit CCD

camera suitable for imaging of FITC-labeled arrays. The images obtained from the

phosphorylated arrays were quantified using the BioNavigator software (PamGene

International BV), and the list of peptides whose phosphorylation was significantly different

between control and test conditions was uploaded to GeneGo for pathway analysis.

Animal experiments. The animal welfare committees of the DIfE as well as the local

authorities (LUGV, Brandenburg, Germany) approved all animal experiments (reference

number V3-2347-37-2011 and 2347-28-2014). All mice were housed in temperature

controlled room (22±1 °C) on a 12:12 h light dark cycle (light on at 6:00 am) and had free

access to food and water at any time. C57BL/6J breeding pairs (Charles River, Germany) as

well as NZO/HIBomDife breeding mice (Dr. Kluge, German Institute of Human Nutrition,

Germany) received a standard chow (SD; V153x R/M-H, Ssniff).

After weaning male C57BL/6J mice were fed a HFD (60 kcal% fat, D12492, Research Diets)

ad libitum in groups of two to six animals per cage. Weekly body weight measurements were

performed in the morning (8-10 am). Classification of mice either prone or resistant to diet-

induced obesity was described before [Kammel et al. 2016 - PMID: 27126637]. At the age of

6 weeks, mice were killed after 6 h fasting. All tissues were directly frozen in liquid nitrogen

and stored at -80 °C until further processing.

Female NZO mice received a standard chow after weaning and starting with 5 weeks of age,

mice were switched to HFD containing 60 kcal% fat (D12492, Research Diets). Classification

of diabetes-resistant (DR) mice in week 10 is based on assessment of liver density by

computed tomography (LaTheta LCT-200, Hitachi-Aloka) combined with the measurement

of blood glucose (CONTOUR® Glucometer, Bayer) [Lubura et al. 2015 - PMID: 26487005].

Liver density <55.2 Hounsfield units and blood glucose values of >8.8 and <16.6 mM both

measured in week 10 predict later onset of diabetes with 83% and 84% probability,

respectively. Combination of both parameters increased the prediction probability to 90 %.

Expression analysis. Total RNA from liver was isolated using QIAzol Lysis Reagent and

RNeasy Mini Kit as recommended. Residual DNA was removed by DNase digestion using

the RNase-Free DNase Set (QIAGEN). Subsequently, cDNA synthesis was conducted with 1

µg RNA, random hexamer primer as well as oligo(dT)15 primer and M-MLV reverse

transcriptase (Promega). qRT-PCR was performed with 12.5 ng cDNA in an Applied

Biosystems 7500 Fast Real-time PCR system with PrimeTime® qPCR Probe Assay (IDT) for

Pdgfa and the GoTaq® Probe qPCR Master Mix (Promega). Data were normalized to the


expression of Eef2 (TaqMan Gene Expression Assay, Life Technologies) as endogenous

control. For genome expression profile either 4x44K or 8x60K whole mouse genome

microarrays (Agilent Technologies) were used.

Immunoblots analysis. Proteins (40 µg) were subjected to SDS-PAGE analysis on 10% gels

and transferred to nitrocellulose membranes. Rabbit polyclonal for total Akt, phospho-Akt

(Ser-473 and Thr-308) were purchased from Cell Signaling. Monoclonal mouse β actin (clone

AC-74; Sigma-Aldrich) was used as loading control.

Primer sequences. Primers for human PDGFA (sense 5′- GACCAGGACGGTCATTTACG -

3′; antisense 5′- CGCACTCCAAATGCTCCT-3′). Primers for mouse Pdgfa (sense 5′-

CAAGACCAGGACGGTCATTT -3′; antisense 5′- GATGGTCTGGGTTCAGGTTG-3′).

Primers for human RPLP0 (sense 5′- ACCTCCTTTTTCCAGGCTTT -3′; antisense 5′-

CCCACTTTGTCTCCAGTCTTG -3′). Primers for human IRS1 (sense 5′-

ACAGGCTTGGGCACGAGT-3′; antisense 5′- AGACCCTCCTCTGGGTAGGA -3′)

Glycogen measurement. Total glycogen in IHH cells was determined by using Abcam

Glycogen assay kit according to the manufacturer’s instruction. The glycogen content was

determined by using fluorometric assay and the fluorescence was measured using plate reader.

Corrections for background glucose were made in all the samples and the corrected

fluorescent readings were applied to standard curve and amount of glycogen in each well was

determined. The glycogen content of each sample was normalized to their respective protein

content and plotted using Graph Pad Prism 5

Materials. The PDGFRα inhibitor Ki11502, human PDGF-AA recombinant, Phorbol 12-

Myristate 13-myristate (PMA) and Metformin were purchased from Sigma-Aldrich. The anti-

PDGF-AA blocking antibodies were from Merckmillipore. The PKC inhibitor Sautrostaurin

was from Selleckchem.

Quantitative PCR. Total RNA was extracted from IHH cells according to the manufacturer’s

protocol (RNeasy Lipid Tissue Kit, Qiagen). The RNA purity and concentration were

determined by RNA Integrity Number (RNA 6000 Nano Kit, 2100 Bioanalyser, Agilent).

Total RNA was transcribed into cDNA as described (17). Each cDNA sample was quantified

by quantitative real-time polymerase chain reaction using the fluorescent TaqMan 5′-nuclease

assays or a BioRad MyiQ Single-Color Real-Time PCR Detection System using the BioRad

iQ SYBR Green Supermix, with 100 nM primers and 1 µl of template per 20 µl of PCR and

an annealing temperature of 60 °C. Gene expression analysis was normalized against Beta-

Glucuronidase (GUSB) expression or 60S acidic ribosomal protein P0 (RPLP0). The primer

sequences are available in the supplemental information.

Western Blotting and ELISA. Cells were scrapped in cold PBS buffer and then cells pellet

was incubated for 30 minutes on ice in the following lysis buffer (20 mM Tris acetate pH 7,

0.27 mM Sucrose, 1% Triton X-100, 1 mM EDTA, 1 mM EGTA, 1 mM DTT) supplemented

with antiproteases and antiphosphatases (Roche, Meylan, France). Cell lysate was centrifuged

15 minutes at 18,000g and supernatant was collected as total proteins. For Western blotting

experiments, 40 µg of total protein extract was separated on 10% SDS-Polyacrylamide gel

and electrically blotted to nitrocellulose membrane. The proteins were detected after an

overnight incubation of the membrane at 4°C with the specific primary antibodies against

AKT (Santa Cruz Biotechnology, dilution 1:1000), PKCƟ (Abcam, dilution 1:1000), PKCε

(Abcam, dilution 1:1000), IRS1 (Cell Signaling Biotechnology, dilution 1:1000), PDGF-AA

(Merck Millipore, dilution 1:1000), αtubulin (Sigma, dilution 1:5000), phospho-AKT (Ser-


473; Cell Signaling Biotechnology, dilution 1:1000), phospho-PKCƟ (Ser-676, Abcam,

dilution 1:1000), phospho-PKCε (Ser-729, Abcam, dilution 1:1000), phospho-INSR (Tyr-972

and Thr-1375, Abcam, dilution 1:1000) in buffer containing 0.1% Tween 20 with either five

percent BSA or five percent milk (for αTubulin). Proteins were visualized with IRDye800 or

IRDye700 (Eurobio) as secondary antibodies. Quantification was performed using the

Odyssey Infrared Imaging System (Eurobio). PDGFA released in the cell supernatant was

quantified by ELISA kit (R &D Systems) according to the manufacturer’s protocol.


Increased hepatic PDGF-AA signaling mediates liver insulin ... · 5/2/2018 · 1 Increased hepatic...

Documents

Transcript of Increased hepatic PDGF-AA signaling mediates liver insulin ... · 5/2/2018 · 1 Increased hepatic...